Online dictionary learning algorithm with periodic updates and its application to image denoising
作者：
Highlights：
• Dictionary learning for sparse modeling improves performance in various applications.
• We introduce a coefficient update procedure for batch and online dictionary learning algorithms.
• We propose a new coefficient updated version of the MOD dictionary learning algorithm.
• We present a new online, periodically coefficient updated dictionary learning algorithm.
• Simulations demonstrate the improvement provided by the coefficient update procedure.
摘要
•Dictionary learning for sparse modeling improves performance in various applications.•We introduce a coefficient update procedure for batch and online dictionary learning algorithms.•We propose a new coefficient updated version of the MOD dictionary learning algorithm.•We present a new online, periodically coefficient updated dictionary learning algorithm.•Simulations demonstrate the improvement provided by the coefficient update procedure.
论文关键词：Dictionary learning,Sparse representation,Online learning,Image denoising
论文评审过程：Available online 10 December 2013.
论文官网地址：https://doi.org/10.1016/j.eswa.2013.11.036
